The ability to comprehend and remember information from text (be it a novel, a newspaper, or an IRS form) is a critical skill for normal adult functioning. In spite of this, we know relatively little about how older adults of varying abilities respond to the demands of different kinds of reading materials. It is clear that the elderly often have difficulty in standard laboratory memory tasks, and that these deficits are, at least in part, due to differences in the quality and degree of initial encoding. Little has been done to investigate the strategies used by the elderly adult in comprehending (or "encoding") text. This is partly due to the fact that methods for studying the online encoding of text have only been developed in recent years. The proposed research is designed to take advantage of these new methodologies to examine the online processing strategies of elderly readers. Specifically, these methods include the accurate measurement of reading time at the level of the word and at the level of the sentence as a function of different text attributes. Text attributes include those at the level of the word (e.g., frequency, length), the level of the sentence (e.g., length in words, number of propositions). or at a broader text level (e.g., narrativity, familiarity). Multiple regression, in which reading time is the dependent variable predicted by these text attributes, may then be used to examine differential effects of text features in controlling processing time. Such a methodology applied to the study of text processing in later adulthood would enable the investigation of the hypothesis that the older adult distributes processing time differently in the comprehension of text. Additionally, the speed of concept recognition during reading will be used to assess age differences in the online representation of the text. An important feature of this research program is that online measures will be related to recall outcome, thus enabling an examination of (1) the relationship between reading strategy and memory ability, and (2) possible age differences in the nature of this relationship. An important working hypothesis of the proposed research program is that processing deficits in later adulthood may necessitate that the elderly adult make use of different reading strategies in order to maintain high levels of comprehension and recall.